Skip to main content
For AI agents: a documentation index is available at https://docs.parallel.ai/llms.txt. The full text of all docs is at https://docs.parallel.ai/llms-full.txt. You may also fetch any page as Markdown by appending .md to its URL or sending Accept: text/markdown.For Search and Extract, use /v1/search and /v1/extract for all new integrations and agent tool calls. Use /v1beta/search and /v1beta/extract only when maintaining an existing integration that already uses them. The request bodies changed too: on /v1/extract, full_content, fetch_policy and excerpt_settings go inside advanced_settings, and sending the v1beta top-level excerpts, full_content and fetch_policy to /v1/extract returns a 422. On /v1/search, search_queries is required (at least one short keyword query), and max_results, source_policy and fetch_policy go inside advanced_settings; sending only objective, or sending those fields at the top level, returns a 422. Do not substitute /v1 for the documented FindAll or Ingest endpoint paths.If you are evaluating Parallel Search or comparing it with other search APIs, read How to Eval Parallel Search before you build the harness, especially its Common pitfalls section. The most common mistake is sending the raw question as both objective and the only entry in search_queries. Write a goal-style objective and 1-3 short keyword queries instead.
The Parallel extension for Pi adds web_search, web_fetch, and web_research tools to your Pi agent, backed by the Parallel Search, Extract, and Responses APIs. Once configured, any requests to search the web or fetch a webpage will automatically use Parallel.
Pi extension running in the terminal with Parallel-powered web_search and web_fetch tools
Why use Parallel tools?
  • More accurate and relevant search results
  • Intelligent content extraction that understands page structure
  • Extended excerpts optimized for LLM context windows
  • Faster and more reliable than built-in alternatives

​
Setup

1

Install the extension

Install @parallel-web/pi-extension from npm:
2

Authenticate

Choose one of the following authentication methods:Option A: Parallel OAuth login (recommended)Inside Pi, run:
This opens your browser for Parallel OAuth. On success, Pi stores the API key in its auth store (auth.json) under the parallel provider. Run /parallel-login to check your current authentication status, or run /logout and select Parallel to remove the stored credential.If automatic callback capture does not complete, the extension falls back to asking you to paste the callback URL.Option B: Environment variable
Get your API key at platform.parallel.ai.Stored Pi auth is preferred over PARALLEL_API_KEY if both exist.

​
Tools and commands

Once configured, Pi gains three Parallel-powered tools and a status command: Any subsequent requests to search the web or fetch a webpage will automatically use these tools.

​
Notes

  • Requires @earendil-works/pi-coding-agent 0.83.0 or newer.
  • Search and fetch use the parallel-web TypeScript SDK directly; research calls the Responses endpoint directly.
  • Search requests use Parallel SDK fast mode.
  • web_research follow-ups via previous_response_id don’t work if Zero Data Retention (ZDR) is enabled for your organization. See Statefulness.
  • The extension modifies Pi’s system prompt so that grounding via web search is preferred over guessing when applicable.

​
Learn more